A Bioreactor System for Magnetic Resonance Microimaging and Spectroscopy of Chondrocytes and Tissue Repair of articular cartilage secondary to either traumatic injury or degenerative joint disease represents an important therapeutic challenge. In spite of significant progress in understanding the pathogenesis of this highly prevalent disease, there are no well-accepted disease-modifying interventions. The development of a flexible and reliable MRI-compatible cartilage hollow fiber bioreactor (HFBR) system for neocartilage growth has the potential to contribute to therapeutic approaches. First, conditions promoting the development of high-quality cartilage from cells can be studied intensively in such a system, which provides full control over exposure of the developing neocartilage to growth factors, substrate composition, dissolved O2 and CO2 concentrations, temperature, and other environmental factors. While in situ development of cartilage from cells, including both chondrocytes and, potentially, bone marrow stromal cells, in an organism will differ in important ways from the bioreactor conditions, in vitro studies will be able to point the way to appropriate conditions for development of functioning neocartilage from cells. Second, growth of high-quality cartilage in the bioreactor may result in a source of tissue for actual transplantation. Finally, and most generally, regardless of the specifics of eventual cartilage repair and regeneration procedures, the ability to monitor tissue quality will be of clear importance. While arthroscopic biopsies provide such data, permitting assessment of the biochemical and histologic state of the tissue, it is clearly more desirable to utilize noninvasive assessment methods. MRI is becoming increasingly accepted as a noninvasive tool for the measurement of cartilage thickness and volume and of localized pathology while the ability of MRI to noninvasively assess cartilage quality is currently a topic of active research. The availability of a highly controllable system for generating cartilage with widely varying properties in a system permitting detailed MRI assessment would represent a clear advance in this effort. Finally, we note that the MRI-compatible bioreactor provides a flexible test-bed for current and future therapeutic agents and interventions. In summary, as a cellular system, the HFBR shares with other 3D culture systems the ability to support the hyaline cartilage type. Thus, one can evaluate the effect of growth conditions and therapeutics on hyaline cartilage tissue rather than fibrocartilage. As a tissue system, the HFBR permits true macroscopic growth. Thus, cell-matrix interactions and the effects of the matrix barrier to substrate delivery and metabolic product efflux are represented much more realistically than in monolayer systems. Finally, as a test bed for growth conditions and agents, the HFBR provides full control of substrate and perfusion conditions. We have successfully demonstrated that cartilage grown from chick sternal cells in the HFBR will develop and maintain the hyaline phenotype; that morphologic measurements with MRI correlate with tissue histology; and that MRI measurements of local T1, T2, diffusion and MT correlate with biochemical assays of collagen, proteoglycans and hydration. Thus, noninvasive MRI measures provide reliable information about cartilage matrix composition. We have further demonstrated that cartilage growth in the HFBR can be modified by introduction of biologically active compounds, and that the correlations between MRI-derived parameters and biochemical results noted above are maintained in spite of the greater dynamic range of tissue characteristics resulting from these interventions. We have also utilized 31P NMR measurements of pH, inorganic phosphate (Pi) and ATP to demonstrate that the developing cartilage in the bioreactor remains metabolically stable over the typical 4 week growth period. A major focus of our work has, in addition, been to demonstrate that MRI measurements of matrix fixed density correlate with measurements of dynamic and equilibrium compressive moduli. The MRI-derived FCD values correlate with S-GAG content but not with collagen content. These correlations were found to persist even in tissue which has undergone development in the presence of chondroitinase, acting as a catabolic agent on matrix proteoglycans. Noninvasive MRI evaluation of FCD therefore has been shown to provide reliable information about cartilage matrix composition under the dynamic conditions of the HFBR in both control tissue and in tissue which has undergone degeneration analogous to that seen in osteoarthritis.